Predictive Sales Data Analysis
Budget: £10 – £15 GBP
I’m sitting on a large set of historical sales figures and need more than just a backward-looking report—I want to know what the numbers will do next. Your task is to take these sales datasets, explore them thoroughly, and build a reliable predictive model that turns patterns into clear, forward-facing insights I can act on.
You’ll start by inspecting the files for any irregularities and preparing them for analysis, then move straight into model selection, training, and validation. Feel free to work in Python (pandas, scikit-learn, Prophet) or R if that’s your comfort zone; SQL for ad-hoc queries and a quick visualization tool such as Tableau or Power BI are welcome additions.
Deliverables should include:
• A clean, well-documented dataset
• Reproducible code/notebooks for the predictive model
• An executive-friendly summary (PDF or slide deck) highlighting drivers, forecasts, and confidence levels
• Optional: an interactive dashboard so I can explore the projections myself
Accuracy and clarity are paramount—I need to understand not only what the forecast says but why it says it. If that sounds like your wheelhouse, let’s talk.
You’ll start by inspecting the files for any irregularities and preparing them for analysis, then move straight into model selection, training, and validation. Feel free to work in Python (pandas, scikit-learn, Prophet) or R if that’s your comfort zone; SQL for ad-hoc queries and a quick visualization tool such as Tableau or Power BI are welcome additions.
Deliverables should include:
• A clean, well-documented dataset
• Reproducible code/notebooks for the predictive model
• An executive-friendly summary (PDF or slide deck) highlighting drivers, forecasts, and confidence levels
• Optional: an interactive dashboard so I can explore the projections myself
Accuracy and clarity are paramount—I need to understand not only what the forecast says but why it says it. If that sounds like your wheelhouse, let’s talk.